The effective integration of electric power producing engines and heat-based process systems that recover waste heat from engine exhausts is limited by the incompatibility of the engine's requirements of nearly constant flow, for maximum power and/or efficiency, and the process needs to modulate inlet flow from the engine exhaust to allow variation of process related fuel flow and thereby heat input required for the process. This ability to reliably modulate process fuel is particularly important when combustion emissions controls are required to meet local regulated limits.
The efficiency of an electric power generation engine is typically limited because of the fundamental thermodynamic cycle that requires some heat to be rejected to a sink during the process. Even the ideal Carnot Cycle has limited efficiency, which depends on the maximum temperature achieved in the cycle versus the minimum, or sink temperature at which heat is rejected. The power output from the cycle is then converted into electricity through various types of generators. While the Carnot cycle requires some heat rejection, and thereby loss of useful output from the fuel, non-ideal engines have even a larger amount of rejected heat. These engines include gas turbines, microturbines, compression ignition engines, spark ignition engines, Stirling cycle, etc. to name a few. Given the waste of fuel energy through rejected heat, it is of high interest to use the engine rejected heat to drive an attached process, such as boilers, water heaters, petroleum process heaters, incinerators, etc. to name a few. By recovering typically wasted heat, the combined fuel utilization, including power and heat, increases substantially. In some cases, this Combined Heat and Power (CHP) approach increases fuel utilization efficiency from 30% to over 80%, or a fuel energy utilization increase of 167%. This has obvious benefits in terms of reductions in fuel cost, fuel use, priority pollutants (e.g. NO, NO2, CO, unburned hydrocarbons) and greenhouse gas emissions (e.g. CO2) relative to useful output. While this approach is very positive, in most applications these advantages are not realized because of the incompatibility of the engine exhaust, and contained waste heat, with the process that attempts to recover the heat. The reasons for this incompatibility are described below. This is then followed by a description of the invention that effectively addresses this incompatibility.
Engines include air compressors and hot gas expanders, joined by combustors (internal or external) that raise temperatures to the needed level to meet cycle power output and efficiency requirements. These compressors and expanders typically use rotating components that operate best at a single design point for either maximum power or maximum efficiency. As the rotating speed changes, the flow throughput changes, which then impacts power and efficiency. This is illustrated by the typical plot of spark ignition engine power output (BHP) and efficiencies related to the inverse of BSFC in the left side of the figure and Thermal, Volumetric, Mechanical efficiencies given in the right side of FIG. 1.
A similar plot is given for a typical compression ignition engine (e.g. diesel engine) in FIG. 2, where fuel BSFC is given, which is related to the inverse of engine efficiency. As shown, for maximum power or maximum efficiency, a specific engine speed is required, with the speed range dependent on the type of engine and scale of the engine. If this speed range is reduced, or in some cases, increased, performance is decreased, which is not desirable for economic return in power generation. Furthermore, in some power generation applications, a synchronous generator is used to produce the correct AC frequency compatible with installed electrical equipment. In these cases, the design point rotating speed for optimal performance is linked with the synchronous speed required to produce the correct AC frequency. This can be accomplished through generator design, as well as gear boxes that can change speed between the engine and generator. In all cases, a fixed design point operating speed and exhaust throughput is desired. This same approach applies to gas turbine, microturbine and other types of continuous combustion power system equipment. FIG. 3 gives the flow rate versus engine speed for a typical gas turbine engine. As shown, as speed is reduced, the throughput flow is significantly reduced. FIGS. 4 and 5 show how both efficiency and power increase with speed and corresponding throughput flow. Of course, maximum speed and throughput is limited by sonic velocity and material strength limits. Therefore, as with other engines, a high speed and fixed throughput near the limits is the preferred operating point for maximum engine power output and efficiency.
As shown, there is a specific design point where gas turbine, engine efficiency and power are attractive. Also, as with spark or compression ignition engines, some of these systems need to operate at a fixed speed and thereby exhaust flow to meet synchronous generator frequency targets for proper grid interconnect. Furthermore, with gas turbine engines that operate at high rotational speeds, there are multiplicities of vibration harmonics of the rotating components that can degrade machine integrity or even destroy the engine Continuous operation at these speeds should be avoided. Once these modes are determined by analysis and testing, engine operating speeds are set to a constant speed to avoid overlap with these undesirable vibration modes. Lastly, gas turbine maximum speed is limited by sonic velocity and material strength.
In summary, for best performance and integrity, and to meet electric power AC frequency requirements in some cases, engines should be operated at a single speed that then produces a relatively fixed exhaust gas flow rate. In a CHP application, this engine operation is often incompatible with thermal load cycling processes and associated burner requirements.
Boiler, water heater, process heater, incinerator and other applications typically need to operate over a range of heat inputs to meet variable thermal load demands from the process of interest. Heat input variations, or required turndown, for boilers, etc., can be as high as 10 to 1, with 8 to 1 highly desired. By having this turndown, the output of the boiler, etc., can be well matched to the process of interest. In typical boilers and other equipment, this turndown capability is met by simultaneously reducing the fuel input and, to the extent possible, required air flow to achieve a relatively consistent fuel-to-air ratio that will then maximize boiler efficiency, as well as burner operability. For example, if only fuel flow was reduced with the air flow fixed, the amount of excess air would greatly increase, up to a factor of nine. In this case, the gas temperature would be low and process conditions may not be met. Also, with a high flow rate of gases into the exhaust per fuel use, the stack heat loss would be high. These characteristics are undesirable. Therefore, it is desired to operate at a nearly constant fuel-to-air ratio as the process load or turndown is varied. This then requires that the air flow, or exhaust from the engine, vary by up to a factor of ten. As noted above, this is incompatible with optimal engine operation, which is to run at a fixed speed and constant exhaust flow. Furthermore, if low boiler emissions at good flame stability is desired, the fixed engine exhaust flow creates further problems. Specifically, emissions control burners need to first reduce flame temperature to reduce NOx emissions that are generated in flame zones where temperatures peak. To control temperature, an inert diluent such as Flue Gas Recirculation (FGR) can be used to suppress the flame temperature and thermal NOx production. However, if FGR is used with conventional non-premixed flames, the NOx control is limited because FGR by itself only controls thermal NOx, not prompt NOx (mainly NO2) that is the dominant NOx at low NOx levels of current interest. Prompt NOx is a strong function of the fuel-to-air mixture ratio, and less sensitive to temperature than thermal NOx. To better address prompt NOx, requires moving from conventional non-premixed flame burner designs, where flame zones operate at the ideal fuel-to-air ratio (i.e. stoichiometric conditions) to those where fuel and air are premixed at conditions other than the stoichiometric ratio, to control the fuel and air mixture ratio within the flame zone. In this approach, the fuel (e.g. natural gas) and air can be premixed, with the excess air beyond that required to consume the fuel acting as a diluent for the flame zone. With this “lean” premixed combustion, the flame zone temperature is also suppressed, reducing thermal NOx. However, prompt NOx is also reduced. FIG. 6 presents a plot of the flame temperature versus the oxygen-to-fuel stoichiometry. As shown, flame temperature will be high at near oxygen stoichiometric conditions of two, with temperature and NOx decreasing as the stoichiometric ratio increases above two and the premixed flame becomes more fuel “lean”.
While the flame zone temperature reduction strategy, by FGR or lean premix, suppresses NOx, it also impacts flame stability. The ability of a local flame to avoid extinction is based on the rate of heat release from fuel oxidation being in balance with the heat loss from the flame zone as a result of contact with cooler gas packets, driven by random turbulent mixing, contact with cooler physical surfaces in the boiler, radiative heat loss with cooler walls and turbulent flame stretch. When kinetically controlled, the heat release is strongly related to flame temperature, which is governed by local mixture ratio, or oxygen stoichiometry, as shown in FIG. 6.
As shown, for either fuel rich (oxygen stoichiometry <2, equivalence ratio >1 or air stoichiometry <1) or fuel lean (oxygen stoichiometry >2, equivalence ratio <1 or air stoichiometry >1) conditions, adiabatic flame temperatures are significantly reduced. In addition, FGR, or higher N2, can ballast and reduce flame temperature, which is positive for NOx emissions, as shown in FIG. 6, but negative for flame stability. With reduced flame temperature for NOx control, the margin between fuel reaction rate and heat loss based extinction is reduced. Therefore, even at the design point condition, NOx emissions control reduces flame integrity and stability. This then requires a more careful control of the fuel-to-air ratio to maintain the proper flame stability margin. Importantly, if the burner requires that the fuel flow is reduced for load reduction, while the air flow is fixed (or nearly fixed) by the engine flow, then the flame will become even leaner as load is reduced. This can be illustrated by a simple example calculation. In the calculation it is assumed that the burner oxidant is a gas turbine exhaust that has an oxygen content of 15%. This flow meets the oxidant requirement for the burner at the full load design point. To achieve the needed NOx, and CO emissions of <9 ppm and <200 ppm, the burner employs fuel rich and fuel lean flame zones that then flow into a single burnout zone. For acceptable flame stability and emissions the rich zone operates at an oxygen stoichiometric ratio of 1.25 or an equivalence ratio of 1.6 and the lean zone operates at an oxygen stoichiometry of 3.22 or an equivalence ratio of 0.62, with the burnout zone operating at an oxygen stoichiometric ratio of 2.32 or an equivalence ratio of 0.86. While the flame stability and emissions control are excellent at full load, as fuel is reduced to match process needs, with a constant exhaust flow, the rich zone becomes less rich and the lean zone becomes leaner. Since both rich and lean zone temperatures are reduced as rich and lean oxygen stoichiometric ratio or equivalence ratios become higher and lower, respectively, as indicated by FIG. 6, then reaction rate falls off and flame stability degrades to the point where flames are extinguished. With a conventional low NOx burner, this is addressed by reducing the oxidant flow as fuel flow is reduced to maintain consistent rich and lean zone equivalence ratios. With the constant engine exhaust flow, in the power burner case, a different and better strategy is required. This is the needed invention, as described below.
This brief background, supports that the constant engine exhaust flow is incompatible with processes that require variable oxidant flow for meeting process requirements and optimizing operating efficiencies as well as meeting currently required emissions regulations.